作者：施莹, 庄哲, 林建辉

作者单位：西南交通大学 牵引动力国家重点实验室, 四川 成都 610031

关键词：高速列车轮对轴承;故障诊断;聚合经验模态分解;排列熵;特征提取;最小二乘支持向量机

摘要：

高速列车轮对轴承的可靠度对高速列车的安全运行具有重要意义，其故障特征主要体现在轴箱振动信号中。该文提出基于聚合经验模态分解排列熵的轮对轴承特征分析方法，提取高速列车轮对轴承振动信号的非线性特征参数，并用于故障状态的分类识别。首先，对高速列车轮对轴箱振动信号进行聚合经验模态分解，得到一系列窄带本征模态函数；然后，对原信号和主要本征模态函数分别计算，得到多组排列熵，形成多尺度的表征信息复杂性高维特征向量；最后，将高维特征向量输入最小二乘支持向量机分类识别出轮对轴承的故障状态。台架试验分析结果表明：该方法针对高速列车轮对轴承故障尤其是轴承复合故障具有较高的识别率，验证通过聚合经验模态分解排列熵对高速列车轮对轴承故障诊断的有效性。

Fault diagnosis method of high speed train axle bearing based on EEMD permutation entropy

SHI Ying, ZHUANG Zhe, LIN Jianhui

State-key Laboratory of Traction Power, Southwest Jiaotong University, Chengdu 610031, China

Abstract: The safety and reliability of high speed train axle bearing is of great significance to the safe operation of high-speed train and its fault feature is mainly reflected in the vibration signal of the axle box. In this paper, a method based on the ensemble empirical mode decomposition of permutation entropy is proposed to extract the nonlinear characteristic parameters of the vibration signal of the high speed train, and used to identify the fault state. Firstly, a series of narrow band intrinsic mode functions are obtained by the empirical mode decomposition of vibration signals. Then, the original signal and the instrinsic mode function are calculated respectively and several groups of permutation entropy are got, and a multi-scale high dimensional feature vector is formed. Finally, the high dimensional feature vector is input to the least squares support vector machine to classify the fault state of the axle bearing. The results of bench test showed that the method has high recognition rate for the axle bearing fault of high-speed train, especially axle bearing compound fault, and the effectives of the method on the high-speed train axle bearing fault diagnosis by ensemble empirical mode decomposition permutation entropy are verified.

Keywords: high speed train axle bearing;fault diagnosis;ensemble empirical mode decomposition;permutation entropy;feature extraction;least squares support vector machine

2017, 43(11): 89-95  收稿日期: 2017-04-23;收到修改稿日期: 2017-05-29

基金项目: 四川省科技计划项目（2016JY0047）

作者简介: 施莹(1987-),女,天津市人,博士研究生,主要从事故障诊断及可靠性方面研究。

参考文献

[1] YANG C Y, WU T Y. Diagnostics of gear deterioration using EEMD approach and PCA process[J]. Measurement,2015(61):75-87.
[2] LOUTRIDIS S J. Damage detection in gear systems using empirical mode decomposition[J]. Engineering,2004(26):1833-1841.
[3] HUANG N E, SHEN Z, LONG S R. The Empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society of London,1998(454):903-995.
[4] 王晗,何刘. 最小熵解卷积法轮对轴承故障诊断[J]. 中国测试,2016,42(1):114-120.
[5] SU W, WANG F, ZHU H,et al. Rolling element bearing faults diagnosis based on optimal Morlet wavelet filter and autocorrelation enhancement[J]. Mechanical Systems & Signal Processing,2010,24(5):1458-1472.
[6] CHENG J S, YU D J, YANG Y. Application of an impulse response wavelet to fault diagnosis of rolling bearings[J]. Mechanical Systems & Signal Processing,2007,21(2):920-929.
[7] YUAN J, HE Z, ZI Y. Gear fault detection using cus-tomized multi wavelet lifting schemes[J]. Mechanical Sy-stems & Signal Processing,2010,24(5):1509-1528.
[8] ZHAO X, PATEL T H, ZUO M J. Multivariate EMD and full spectrum based condition monitoring for rotating machinery[J]. Mechanical Systems & Signal Processing,2012,27(1):712-728.
[9] WU T Y, CHUNG Y L. Misalignment diagnosis of ro-tating machinery through vibration analysis via the hybrid EEMD and EMD approach[J]. Smart Materials & Structures,2009,18(9):7566-7579.
[10] LEI Y, HE Z, ZI Y. EEMD method and WNN for fault diagnosis of locomotive roller bearings[J]. Expert Systems with Applications An International Journal,2011,38(6):7334-7341.
[11]ŽVOKELJ M, ZUPAN S, PREBIL I. Non-linear multi-variate and multiscale monitoring and signal denoising strategy using kernel principal component analysis combined with ensemble empirical mode decomposition method[J]. Mechanical Systems & Signal Processing,2011, 25(7):2631-2653.
[12] 秦娜,金炜东,黄进,等. 高速列车转向架故障信号的聚合经验模态分解和模糊熵特征分析[J]. 控制理论与应用,2014,31(9):1245-1251.
[13] BANDT C, POMPE B. Permutation entropy:a natural complexity measure for time series[J]. Physical Review Letters,2002,88(17):174102.
[14] 冯辅周,司爱威,饶国强,等. 基于小波相关排列熵的轴承早期故障诊断技术[J]. 机械工程学报,2012,48(13):73-79.
[15] ZUNINO L, ROSSO O A, SORIANO M C. Character-izing the hyperchaotic dynamics of a semiconductor laser subject to optical feedback via permutation entropy[J]. IEEE Journal of Selected Topics in Quantum Electronics, 2011,17(5):1250-1257.
[16] 秦娜,蒋鹏,孙永奎,等. 基于EEMD排列熵的高速列车转向架故障特征分析[J]. 振动、测试与诊断,2015,35(5):885-891.
[17] 易彩. 高速列车轮对轴承状态表征与故障诊断方法研究[D]. 成都:西南交通大学,2015.
[18] 刘永斌,龙潜,冯志华,等. 一种非平稳、非线性振动信号检测方法的研究[J]. 振动与冲击,2007,26(12):131-134.